Dynamic Scoring: A novel method for quantitative modeling of guest-host associati

描述（由申请人提供）：本提案的目标是开发一种新的计算方法，以明确考虑蛋白质灵活性的方式有效量化蛋白质-配体缔合。通过非共价缔合的分子之间的分子识别在生物系统中的几乎所有过程中起着基本作用。虽然存在许多计算概念来模拟药物-蛋白质识别，但这些相互作用的有效和准确的定量仍然没有实现。我们提出了一种新的计算方法，解决了一些最严重的缺点，目前的方法：蛋白质的灵活性和可靠的定量结合亲和力。我们引入了一个假设的“配体模型”的新概念：一个虚拟的配体，结合到蛋白质和动态改变其形状和性质在分子动力学（MD）模拟，基本上代表了一个大的合奏不同的化学物种结合到同一个目标蛋白质。这种方法允许对与其与化学品或候选药物相互作用相关的蛋白质构象进行采样。这种方法也将允许我们探测配体结合后蛋白质的构象灵活性。“配体模型”的概念将导致一个有效的解耦采样使用MD模拟和随后的对接。因此，该方法结合了定量分子识别的准确性和大型化合物文库虚拟筛选的效率。预计该软件将引起蛋白质-配体相互作用所有领域研究人员的广泛兴趣，包括药物设计，结构生物学和环境毒理学。 公共卫生相关性分子间通过非共价缔合的分子识别在生物系统的几乎所有过程中起着基本作用。该项目旨在开发一种新的计算方法来有效地量化蛋白质-配体结合，明确包括蛋白质的动力学。它将在药物设计和环境毒理学方面具有广泛的适用性。

DESCRIPTION (provided by applicant): The goal of this proposal is to develop a new computational method to efficiently quantify protein-ligand association in a way that explicitly considers protein flexibility. Molecular recognition between molecules through noncovalent association plays a fundamental role in virtually all processes in biological systems. Although many computational concepts exist to simulate drug-protein recognition, an efficient and accurate quantification of these interactions has still not been achieved. We propose a novel computational method that addresses some of the most serious shortcomings of present approaches: protein flexibility and a reliable quantification of binding affinities. We introduce the new concept of a hypothetical 'ligand model': a virtual ligand that binds to the protein and dynamically changes its shape and properties during molecular dynamics (MD) simulations, essentially representing a large ensemble of different chemical species binding to the same target protein. This approach allows sampling protein conformations relevant to its interaction with chemicals or drug candidates. This method also will allow us to probe conformational flexibility of the protein upon ligand binding. The 'ligand-model' concept will result in an efficient decoupling of sampling using MD simulations and subsequent docking. This method consequently combines both accuracy in quantifying molecular recognition and efficiency in virtual screening of large compound libraries. The software is anticipated to be of wide interest for researchers in all areas of protein-ligand interactions, including drug design, structural biology, and environmental toxicology. PUBLIC HEALTH RELEVANCE Molecular recognition between molecules through noncovalent association plays a fundamental role in virtually all processes in biological systems. This project is aimed toward developing a novel computational method to efficiently quantify protein-ligand binding, explicitly including the dynamics of the protein. It will have wide applicability for drug design and environmental toxicology.